Prepare for the PSP by training process judgment, not software clicks. Master the TCM planning and scheduling process, compute CPM passes by hand, reason about logic versus constraints, interpret updates against a data date, and choose between time impact analysis, windows analysis, and as-built comparison for a stated question.
Why the PSP rewards process thinking over software mechanics
The PSP credential centers on leading the planning and scheduling process within AACE's TCM framework: developing, monitoring, updating, forecasting, and analyzing integrated project schedules and communicating them to stakeholders.
Treat planning and scheduling as two distinct disciplines. Planning decides how the work will be executed: scope breakdown, sequencing strategy, durations, resources, and assumptions. Scheduling expresses that plan as a dated, logic-driven model and maintains it. When you study, narrate every technique in those terms: a WBS decomposition is a planning act; recalculating float after a data date is a scheduling act.
Anchor your review to the TCM planning and scheduling process rather than to software menus. Walk the lifecycle in order: develop the planning basis, build the schedule model, baseline it, status and update it against progress, analyze variances and trends, and forecast. For each stage, write one sentence on what inputs it consumes and what deliverable it produces. That lifecycle map is the spine on which every other topic in this guide hangs.
Planning outputs: WBS decomposition, activity definition, and the schedule basis
Planning turns scope into schedule-ready work definitions. Know how a WBS differs from the activity list, what milestones and hammocks represent, and why the schedule basis document records the assumptions behind every duration and logic choice.
Trace the decomposition path explicitly. The WBS organizes deliverables and work into a hierarchical structure for management control; the activity list then breaks work packages into schedulable tasks with durations, logic, and resources. These are different artifacts serving different purposes: WBS elements roll up cost and scope, while activities consume time and drive the network. Confusing them produces schedules that report cost at the wrong level or logic at the wrong granularity.
Study the supporting artifacts with the same care as the network. Milestones mark significant events with zero duration; hammocks summarize grouped work over a time span. The schedule basis document captures calendars, assumptions, exclusions, and logic rationale so the model is auditable. Practice writing a three-sentence basis note for any schedule you build: what the plan assumes, what it excludes, and why the controlling logic was chosen. Building that documentation habit keeps the scenario exercises in this guide grounded in justification rather than arithmetic alone.
The CPM forward and backward pass, computed by hand
The critical path method drives every schedule question. Compute the forward pass for early dates, the backward pass for late dates, and read total float and free float directly from the results before touching any tool.
Work a small network until the mechanics are automatic. Suppose A (5 days) feeds both B (8 days) and C (6 days), and both feed D (4 days). Forward pass: A runs days 0-5; B runs 5-13; C runs 5-11; D starts at 13 and finishes at 17. Backward pass: D must finish by 17, so its late start is 13; B then has late dates 5-13 and zero float; C has late dates 7-13. This takes two minutes on paper and exposes the structure behind every forecast.
Now read the float definitions off the numbers. Total float is late finish minus early finish: C has 2 days, B has 0. Free float is how long an activity can slip without delaying its earliest successor: C can slip to day 13, so its free float is also 2 here; B's is 0 because D's early start equals B's early finish. On networks where paths merge, total float and free float diverge, and knowing which one a scenario question is asking about changes the answer. Practice until you can state, for any activity, which delay hurts the project and which only hurts the next crew.
Logic types, lags, and when a constraint corrupts your float
Finish-to-start is the default logic; start-to-start and finish-to-finish with lags model overlapping work. Constraints pin dates regardless of logic, so substituting a constraint for a missing relationship distorts float and hides gaps in the network.
Scenario: a roofing subcontractor cannot start until the top-floor structural inspection passes, but your scheduler enters a start-no-earlier-than date for the roofing instead of tying it to the inspection milestone. The plausible mistake is treating the constraint as a shortcut. The better decision is to connect roofing to the inspection milestone with finish-to-start logic, or use a start-to-start lag only where a genuine overlap exists. Why it matters: the constraint calculates a date that looks reasonable today, but it severs the causal link, so the roofing shows misleading float, the inspection can slip without pushing roofing, and any quality check for open logic will not flag the real gap.
Learn the relationship vocabulary precisely. Finish-to-start says a successor cannot finish before its predecessor finishes; start-to-start and finish-to-finish tie starts to starts and finishes to finishes, usually with lags for procurement lead time or curing time. Constraints come in flavors such as early, late, and mandatory, and each overrides a different part of the calculation. In scenarios, ask two questions before accepting any constraint: what real-world event does it represent, and is that event better modeled as logic? A well-built schedule lets logic, not pinned dates, carry the sequence so float remains a truthful measure of flexibility.
Data dates, updates, and out-of-sequence progress decisions
An update statuses progress as of the data date and forecasts remaining work. Out-of-sequence completions force a modeling choice about how the engine treats progressed logic, and that choice changes the remaining forecast.
Scenario: at the monthly update, a mechanical activity is marked complete even though its predecessor, the ceiling closeout, is still open. The scheduler clicks update and ships the forecast. The plausible mistake is accepting the tool's default handling of out-of-sequence progress without checking it. The better decision is to recognize the situation, decide deliberately between treating remaining logic as retained or overridden, and document the selection in the schedule basis. Why it matters: the two conventions can produce different remaining durations and different critical paths for the identical progress data, so an undocumented default is an unexplained forecast.
Build your update literacy around the data date, the point in time separating actuals from forecasts. Every statused activity contributes actual dates and remaining durations; everything after the data date is a forward calculation. Distinguish an update, which records progress, from a schedule analysis, which interprets it: float erosion, critical path migration, and trend against the baseline. Practice reading an update report in that order: what changed since the last data date, what does the longest path look like now, and which assumptions in the basis no longer hold. That reading discipline is the interpretive skill the update scenarios in this guide exercise.
Choosing between analysis methods: TIA, windows, and as-built comparison
Each schedule analysis method answers a different question. Match the method to the question and the available records: a proposed change calls for time impact analysis, retrospective delay attribution calls for windows analysis, and diagnosis starts with as-planned versus as-built.
Train method selection as a matching exercise. Time impact analysis inserts a fragnet, a small network fragment representing the changed or delayed work, into the current schedule to measure a single event's effect on the forecast. Windows analysis divides the project into reporting periods and asks which activities drove delay within each window. As-planned versus as-built comparison lays the baseline beside recorded actual dates to see where execution diverged. Each consumes different evidence and supports a different kind of conclusion, and that discrimination is what the scenario exercises in this guide train.
Calibrate each method to its assumptions rather than treating outputs as universal truth. A time impact analysis is only as good as the fragnet's logic and the currency of the schedule it is inserted into. A windows analysis depends on periodic updates actually existing and being reliable. An as-built critical path traced through recorded dates is retrospective evidence, not a plan, and it assumes the recorded sequence is complete. Practice writing one sentence per method stating its core assumption; that habit keeps your answers conditional instead of overclaiming.
Use the table below as a decision aid during review. Cover the right-hand columns, read each method name, and recite the question it answers and the inputs it requires.
| Method | Question it answers | Inputs required | Best-fit context |
|---|---|---|---|
| Time impact analysis (TIA) | How does this specific change or delay event move the forecast finish? | Current schedule update plus a fragnet modeling the event | Evaluating a proposed change or a discrete delay event prospectively |
| Windows analysis | Which activities drove delay during each reporting period? | A series of periodic updates spanning the delay period | Retrospective delay attribution across a project's life |
| As-planned vs. as-built comparison | Where and how did actual execution diverge from the baseline plan? | Baseline schedule and recorded actual start and finish dates | Early diagnostic step before committing to deeper analysis |
| As-built critical path | Which actual sequence of work controlled completion? | Complete as-built dates and a reconstructed actual logic network | Retrospective review where periodic updates are missing or unreliable |
A preparation sequence, self-check rubric, and readiness checks
Sequence your study from TCM process to hand mechanics to scenario judgment, and measure readiness with a rubric you can score honestly rather than with hours logged.
Practical exercise: draw a five-activity network with two merging paths, compute the forward and backward pass by hand, and record each activity's total float and free float. Then insert one start-no-earlier-than constraint on a non-critical activity and recalculate. Expected observations: the constrained activity's float collapses toward zero even though its real flexibility did not change, its predecessor relationships stop influencing its dates, and the critical path may migrate for reasons that have nothing to do with actual work. Score yourself on this rubric, one point each: passes computed correctly; total and free float correctly distinguished on a merging activity; the constraint's float effect described in one sentence; a three-sentence schedule basis note written for the network; a retained-logic versus progress-override explanation given without notes. Five out of five is a learning milestone, not a passing prediction.
An adaptable eight-block sequence: blocks one and two, read the TCM planning and scheduling process and build your lifecycle map with the artifacts from section two; blocks three and four, drill hand-computed passes and float until section-three style networks take minutes; block five, work logic-versus-constraint scenarios like section four; block six, practice reading updates and out-of-sequence situations like section five; block seven, drill the analysis-method matching table until selection is reflexive; block eight, run mixed scenarios and score them against the rubric, then use practice questions to find the concepts, not just the answers, that still wobble. Readiness checks: you can compute passes on paper, justify any constraint in logic terms, explain what a data date separates, pick an analysis method for a stated question, and outline a schedule basis document. For administrative matters such as eligibility, exam format, and delivery, confirm current requirements directly with AACE International rather than relying on secondary summaries.
- Rubric line 1: forward and backward pass computed correctly on a merging-path network.
- Rubric line 2: total float and free float distinguished for an activity whose paths merge downstream.
- Rubric line 3: a constraint's distortion of float explained in one written sentence.
- Rubric line 4: a three-sentence schedule basis note covering assumptions, exclusions, and controlling logic.
- Rubric line 5: retained logic versus progress override explained without reference notes.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.